{"url":"/sota/lidar-semantic-segmentation-on-nuscenes","task":{"name":"LIDAR Semantic Segmentation","url":"/task/lidar-semantic-segmentation","note":null},"dataset":{"name":"nuScenes","url":"/dataset/nuscenes"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":null,"description_from":null,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["test mIoU","val mIoU"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"test mIoU":null,"val mIoU":null}},"counts":{"rows":36,"rows_with_code":19,"rows_with_paper_page":26,"rows_dated":26,"rows_using_additional_data":2},"rows":[{"rank_in_archive_order":1,"model":"DITR","metrics":{"test mIoU":"0.851","val mIoU":"0.842"},"uses_additional_data":false,"paper_date":"2025-03-24","paper":"/paper/dino-in-the-room-leveraging-2d-foundation","paper_url":"https://arxiv.org/abs/2503.18944v1","paper_title":"DINO in the Room: Leveraging 2D Foundation Models for 3D Segmentation","code":"https://github.com/VisualComputingInstitute/DITR","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"PTv3 + PPT","metrics":{"test mIoU":"0.830","val mIoU":"0.812"},"uses_additional_data":true,"paper_date":"2023-12-15","paper":"/paper/point-transformer-v3-simpler-faster-stronger","paper_url":"https://arxiv.org/abs/2312.10035v2","paper_title":"Point Transformer V3: Simpler, Faster, Stronger","code":"https://github.com/Pointcept/Pointcept","n_code_links":3,"syntology":{"n_ran":7,"n_unverified":4,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"PTv2","metrics":{"test mIoU":"0.826","val mIoU":"0.802"},"uses_additional_data":false,"paper_date":"2022-10-11","paper":"/paper/point-transformer-v2-grouped-vector-attention","paper_url":"https://arxiv.org/abs/2210.05666v2","paper_title":"Point Transformer V2: Grouped Vector Attention and Partition-based Pooling","code":"https://github.com/Pointcept/Pointcept","n_code_links":2,"syntology":null},{"rank_in_archive_order":4,"model":"FRNet","metrics":{"test mIoU":"0.825","val mIoU":"0.790"},"uses_additional_data":false,"paper_date":"2023-12-07","paper":"/paper/frnet-frustum-range-networks-for-scalable","paper_url":"https://arxiv.org/abs/2312.04484v3","paper_title":"FRNet: Frustum-Range Networks for Scalable LiDAR Segmentation","code":"https://github.com/ldkong1205/Robo3D","n_code_links":2,"syntology":{"n_ran":14,"n_unverified":5,"n_samples":19,"n_pointer_only_licence":18}},{"rank_in_archive_order":5,"model":"SphereFormer","metrics":{"test mIoU":"0.819","val mIoU":"0.795"},"uses_additional_data":false,"paper_date":"2023-03-22","paper":"/paper/spherical-transformer-for-lidar-based-3d","paper_url":"https://arxiv.org/abs/2303.12766v1","paper_title":"Spherical Transformer for LiDAR-based 3D Recognition","code":"https://github.com/dvlab-research/sphereformer","n_code_links":2,"syntology":{"n_ran":8,"n_unverified":5,"n_samples":13,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"SPVCNN++","metrics":{"test mIoU":"0.811"},"uses_additional_data":false,"paper_date":"2020-07-31","paper":"/paper/searching-efficient-3d-architectures-with","paper_url":"https://arxiv.org/abs/2007.16100v2","paper_title":"Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution","code":"https://github.com/Pointcept/Pointcept","n_code_links":6,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"2DPASS","metrics":{"test mIoU":"0.81"},"uses_additional_data":false,"paper_date":"2022-07-10","paper":"/paper/2dpass-2d-priors-assisted-semantic","paper_url":"https://arxiv.org/abs/2207.04397v3","paper_title":"2DPASS: 2D Priors Assisted Semantic Segmentation on LiDAR Point Clouds","code":"https://github.com/yanx27/2dpass","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"RangeFormer","metrics":{"test mIoU":"0.81"},"uses_additional_data":false,"paper_date":"2023-03-09","paper":"/paper/rethinking-range-view-representation-for","paper_url":"https://arxiv.org/abs/2303.05367v3","paper_title":"Rethinking Range View Representation for LiDAR Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":9,"model":"SFPNet","metrics":{"test mIoU":"0.802","val mIoU":"0.801"},"uses_additional_data":false,"paper_date":"2024-07-16","paper":"/paper/sfpnet-sparse-focal-point-network-for","paper_url":"https://arxiv.org/abs/2407.11569v1","paper_title":"SFPNet: Sparse Focal Point Network for Semantic Segmentation on General LiDAR Point Clouds","code":"https://github.com/Cavendish518/SFPNet","n_code_links":1,"syntology":{"n_ran":7,"n_unverified":3,"n_samples":10,"n_pointer_only_licence":10}},{"rank_in_archive_order":10,"model":"2D3DNet","metrics":{"test mIoU":"0.80"},"uses_additional_data":false,"paper_date":"2021-10-21","paper":"/paper/learning-3d-semantic-segmentation-with-only","paper_url":"https://arxiv.org/abs/2110.11325v1","paper_title":"Learning 3D Semantic Segmentation with only 2D Image Supervision","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":11,"model":"GU-Net","metrics":{"test mIoU":"0.80"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":12,"model":"DRINet++: Efficient Voxel-as-point Point Cloud Segmentation","metrics":{"test mIoU":"0.80"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":13,"model":"AF2S3Net","metrics":{"test mIoU":"0.78"},"uses_additional_data":false,"paper_date":"2021-02-08","paper":"/paper/af-2-s3net-attentive-feature-fusion-with","paper_url":"https://arxiv.org/abs/2102.04530v1","paper_title":"(AF)2-S3Net: Attentive Feature Fusion with Adaptive Feature Selection for Sparse Semantic Segmentation Network","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":14,"model":"Cylinder3D++","metrics":{"test mIoU":"0.78"},"uses_additional_data":false,"paper_date":"2020-08-04","paper":"/paper/cylinder3d-an-effective-3d-framework-for","paper_url":"https://arxiv.org/abs/2008.01550v1","paper_title":"Cylinder3D: An Effective 3D Framework for Driving-scene LiDAR Semantic Segmentation","code":"https://github.com/xinge008/Cylinder3D","n_code_links":3,"syntology":null},{"rank_in_archive_order":15,"model":"CPFusion","metrics":{"test mIoU":"0.78"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":16,"model":"SPVNAS","metrics":{"test mIoU":"0.77"},"uses_additional_data":false,"paper_date":"2020-07-31","paper":"/paper/searching-efficient-3d-architectures-with","paper_url":"https://arxiv.org/abs/2007.16100v2","paper_title":"Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution","code":"https://github.com/Pointcept/Pointcept","n_code_links":6,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":17,"model":"AMVNet","metrics":{"test mIoU":"0.77"},"uses_additional_data":false,"paper_date":"2020-12-09","paper":"/paper/amvnet-assertion-based-multi-view-fusion","paper_url":"https://arxiv.org/abs/2012.04934v1","paper_title":"AMVNet: Assertion-based Multi-View Fusion Network for LiDAR Semantic Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":18,"model":"Cylinder3D+InstanceAug","metrics":{"test mIoU":"0.77"},"uses_additional_data":false,"paper_date":"2020-11-19","paper":"/paper/cylindrical-and-asymmetrical-3d-convolution","paper_url":"https://arxiv.org/abs/2011.10033v1","paper_title":"Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation","code":"https://github.com/xinge008/Cylinder3D","n_code_links":2,"syntology":null},{"rank_in_archive_order":19,"model":"PMF-ResNet50","metrics":{"test mIoU":"0.77"},"uses_additional_data":false,"paper_date":"2021-06-21","paper":"/paper/perception-aware-multi-sensor-fusion-for-3d","paper_url":"https://arxiv.org/abs/2106.15277v3","paper_title":"EPMF: Efficient Perception-aware Multi-sensor Fusion for 3D Semantic Segmentation","code":"https://github.com/ICEORY/PMF","n_code_links":1,"syntology":null},{"rank_in_archive_order":20,"model":"LIFusion","metrics":{"test mIoU":"0.76"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":21,"model":"GFNet","metrics":{"test mIoU":"0.76"},"uses_additional_data":false,"paper_date":"2022-07-06","paper":"/paper/gfnet-geometric-flow-network-for-3d-point","paper_url":"https://arxiv.org/abs/2207.02605v2","paper_title":"GFNet: Geometric Flow Network for 3D Point Cloud Semantic Segmentation","code":"https://github.com/haibo-qiu/gfnet","n_code_links":1,"syntology":null},{"rank_in_archive_order":22,"model":"OrangeNet","metrics":{"test mIoU":"0.75"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":23,"model":"JS3C-Net","metrics":{"test mIoU":"0.74"},"uses_additional_data":false,"paper_date":"2020-12-07","paper":"/paper/sparse-single-sweep-lidar-point-cloud","paper_url":"https://arxiv.org/abs/2012.03762v1","paper_title":"Sparse Single Sweep LiDAR Point Cloud Segmentation via Learning Contextual Shape Priors from Scene Completion","code":"https://github.com/yanx27/JS3C-Net","n_code_links":2,"syntology":null},{"rank_in_archive_order":24,"model":"Open_world_incremental","metrics":{"test mIoU":"0.74"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":25,"model":"PolarStream-1","metrics":{"test mIoU":"0.73"},"uses_additional_data":false,"paper_date":"2021-06-14","paper":"/paper/polarstream-streaming-lidar-object-detection","paper_url":"https://arxiv.org/abs/2106.07545v2","paper_title":"PolarStream: Streaming Lidar Object Detection and Segmentation with Polar Pillars","code":null,"n_code_links":0,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":26,"model":"DB-Unet","metrics":{"test mIoU":"0.70"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":27,"model":"Vis-PolarNet","metrics":{"test mIoU":"0.70"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":28,"model":"PolarNet","metrics":{"test mIoU":"0.69"},"uses_additional_data":false,"paper_date":"2020-03-31","paper":"/paper/polarnet-an-improved-grid-representation-for","paper_url":"https://arxiv.org/abs/2003.14032v2","paper_title":"PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation","code":"https://github.com/edwardzhou130/PolarSeg","n_code_links":4,"syntology":{"n_ran":11,"n_unverified":0,"n_samples":11,"n_pointer_only_licence":1}},{"rank_in_archive_order":29,"model":"EfficientLPS","metrics":{"test mIoU":"0.68"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":30,"model":"Camera-LiDAR Fusion + HTC + Map","metrics":{"test mIoU":"0.44"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":31,"model":"Serialized Piont Mamba","metrics":{"val mIoU":"0.806"},"uses_additional_data":false,"paper_date":"2024-07-17","paper":"/paper/serialized-point-mamba-a-serialized-point","paper_url":"https://arxiv.org/abs/2407.12319v1","paper_title":"Serialized Point Mamba: A Serialized Point Cloud Mamba Segmentation Model","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":32,"model":"LSK3DNet","metrics":{"val mIoU":"0.801"},"uses_additional_data":false,"paper_date":"2024-03-22","paper":"/paper/lsk3dnet-towards-effective-and-efficient-3d","paper_url":"https://arxiv.org/abs/2403.15173v1","paper_title":"LSK3DNet: Towards Effective and Efficient 3D Perception with Large Sparse Kernels","code":"https://github.com/fengzicai/lsk3dnet","n_code_links":1,"syntology":null},{"rank_in_archive_order":33,"model":"WaffleIron","metrics":{"val mIoU":"0.791"},"uses_additional_data":false,"paper_date":"2023-01-24","paper":"/paper/using-a-waffle-iron-for-automotive-point","paper_url":"https://arxiv.org/abs/2301.10100v2","paper_title":"Using a Waffle Iron for Automotive Point Cloud Semantic Segmentation","code":"https://github.com/valeoai/waffleiron","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":0,"n_samples":5,"n_pointer_only_licence":5}},{"rank_in_archive_order":34,"model":"OA-CNNs","metrics":{"val mIoU":"0.789"},"uses_additional_data":false,"paper_date":"2024-03-21","paper":"/paper/oa-cnns-omni-adaptive-sparse-cnns-for-3d","paper_url":"https://arxiv.org/abs/2403.14418v1","paper_title":"OA-CNNs: Omni-Adaptive Sparse CNNs for 3D Semantic Segmentation","code":"https://github.com/Pointcept/Pointcept","n_code_links":1,"syntology":null},{"rank_in_archive_order":35,"model":"PPT+SparseUNet","metrics":{"val mIoU":"0.786"},"uses_additional_data":true,"paper_date":"2023-08-18","paper":"/paper/towards-large-scale-3d-representation","paper_url":"https://arxiv.org/abs/2308.09718v2","paper_title":"Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Training","code":"https://github.com/Pointcept/Pointcept","n_code_links":1,"syntology":null},{"rank_in_archive_order":36,"model":"Point-to-Voxel KD","metrics":{"val mIoU":"0.760"},"uses_additional_data":false,"paper_date":"2022-06-05","paper":"/paper/point-to-voxel-knowledge-distillation-for-1","paper_url":"https://arxiv.org/abs/2206.02099v1","paper_title":"Point-to-Voxel Knowledge Distillation for LiDAR Semantic Segmentation","code":null,"n_code_links":0,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,885 of the 9,623 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9623,"papers_checked":6885,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2737},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-25T09:33:49+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":9,"rows_with_any_sample_ran":7,"distinct_papers_with_graph_line":8,"distinct_papers_with_any_sample_ran":7,"samples_over_distinct_papers":{"n_ran":53,"n_unverified":20,"n_samples":73,"n_pointer_only_licence":34,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":53,"n_unverified":23,"n_samples":76,"n_pointer_only_licence":34,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}